Real-Time and Data-Driven Operation Optimization and Knowledge Discovery for an Enterprise Information System

An enterprise information system (EIS) is an integrated data-applications platform characterized by diverse, heterogeneous, and distributed data sources. For many enterprises, a number of business processes still depend heavily on static rule-based methods and extensive human expertise. Enterprises are faced with the need for optimizing operation scheduling, improving resource utilization, discovering useful knowledge, and making data-driven decisions.

This thesis research is focused on real-time optimization and knowledge discovery that addresses workflow optimization, resource allocation, as well as data-driven predictions of process-execution times, order fulfillment, and enterprise service-level performance. In contrast to prior work on data analytics techniques for enterprise performance optimization, the emphasis here is on realizing scalable and real-time enterprise intelligence based on a combination of heterogeneous system simulation, combinatorial optimization, machine-learning algorithms, and statistical methods.

On-demand digital-print service is a representative enterprise requiring a powerful EIS.We use real-life data from Reischling Press, Inc. (RPI), a digit-print-service provider (PSP), to evaluate our optimization algorithms.

In order to handle the increase in volume and diversity of demands, we first present a high-performance, scalable, and real-time production scheduling algorithm for production automation based on an incremental genetic algorithm (IGA). The objective of this algorithm is to optimize the order dispatching sequence and balance resource utilization. Compared to prior work, this solution is scalable for a high volume of orders and it provides fast scheduling solutions for orders that require complex fulfillment procedures. Experimental results highlight its potential benefit in reducing production inefficiencies and enhancing the productivity of an enterprise.

We next discuss analysis and prediction of different attributes involved in hierarchical components of an enterprise. We start from a study of the fundamental processes related to real-time prediction. Our process-execution time and process status prediction models integrate statistical methods with machine-learning algorithms. In addition to improved prediction accuracy compared to stand-alone machine-learning algorithms, it also performs a probabilistic estimation of the predicted status. An order generally consists of multiple series and parallel processes. We next introduce an order-fulfillment prediction model that combines advantages of multiple classification models by incorporating flexible decision-integration mechanisms. Experimental results show that adopting due dates recommended by the model can significantly reduce enterprise late-delivery ratio. Finally, we investigate service-level attributes that reflect the overall performance of an enterprise. We analyze and decompose time-series data into different components according to their hierarchical periodic nature, perform correlation analysis,

and develop univariate prediction models for each component as well as multivariate models for correlated components. Predictions for the original time series are aggregated from the predictions of its components. In addition to a significant increase in mid-term prediction accuracy, this distributed modeling strategy also improves short-term time-series prediction accuracy.

In summary, this thesis research has led to a set of characterization, optimization, and prediction tools for an EIS to derive insightful knowledge from data and use them as guidance for production management. It is expected to provide solutions for enterprises to increase reconfigurability, accomplish more automated procedures, and obtain data-driven recommendations or effective decisions.

Genetic Studies Identify Critical Biomarkers and Refine the Classification of Malignant Gliomas

Gliomagenesis is driven by a complex network of genetic alterations and while the glioma genome has been a focus of investigation for many years; critical gaps in our knowledge of this disease remain. The identification of novel molecular biomarkers remains a focus of the greater cancer community as a method to improve the consistency and accuracy of pathological diagnosis. In addition, novel molecular biomarkers are drastically needed for the identification of targets that may ultimately result in novel therapeutics aimed at improving glioma treatment. Through the identification of new biomarkers, laboratories will focus future studies on the molecular mechanisms that underlie glioma development. Here, we report a series of genomic analyses identifying novel molecular biomarkers in multiple histopathological subtypes of glioma and refine the classification of malignant gliomas. We have completed a large scale analysis of the WHO grade II-III astrocytoma exome and report frequent mutations in the chromatin modifier, alpha thalassemia mental retardation x-linked (ATRX), isocitrate dehydrogenase 1 and 2 (IDH1 and IDH2), and mutations in tumor protein 53 (TP53) as the most frequent genetic mutations in low grade astrocytomas. Furthermore, by analyzing the status of recurrently mutated genes in 363 brain tumors, we establish that highly recurrent gene mutational signatures are an effective tool in stratifying homogeneous patient populations into distinct groups with varying outcomes, thereby capable of predicting prognosis. Next, we have established mutations in the promoter of telomerase reverse transcriptase (TERT) as a frequent genetic event in gliomas and in tissues with low rates of self renewal. We identify TERT promoter mutations as the most frequently mutated gene in primary glioblastoma. Additionally, we show that TERT promoter mutations in combination with IDH1 and IDH2 mutations are able to delineate distinct clinical tumor cohorts and are capable of predicting median overall survival more effectively than standard histopathological diagnosis alone. Taken together, these data advance our understanding of the genetic alterations that underlie the transformation of glial cells into neoplasms and we provide novel genetic biomarkers and multi – gene mutational signatures that can be utilized to refine the classification of malignant gliomas and provide opportunity for improved diagnosis.